Now you can look at one of them

I am starting a long project. My wife and I are going to sift through, process, and organize everything we have kept: the last fifteen to twenty years of my life, and hers, which together span something like thirty to forty years of documents, files, photos, videos, code, old projects, websites, and paper. All of it. This post is the first in a series about doing that, and about why.

Why

When people look back at a period, they aggregate. They take millions of lives, average them, and write a sentence: people of that time did x, y, and z. The sentence is usually true and almost entirely empty. It tells you what a generation did, but not what it was like to be one person inside it, with one desk, one inbox, one shoebox of receipts, and one hard drive that nearly died in 2011.

I want there to be an answer to that. When someone says “people of our time did x, y, and z,” I want it to be possible to reply: now you can look at one of them. Not a summary, not a highlight reel, but the whole thing, with the boring parts left in. The tax forms next to the vacation photos. The abandoned side project next to the one that shipped. The website I made at nineteen that I would rather nobody saw, kept anyway, because the embarrassment is part of the record.

This is record keeping for posterity, in the plain sense of the word. Not for an audience I can name. For whoever comes later and is curious.

What we have

Between the two of us, roughly:

• Paper. Letters, bills, contracts, school work, notebooks, certificates, tickets, receipts, printed photos, and the folders and boxes they live in.
• Documents. Decades of word processor files, spreadsheets, PDFs, scans, and exports from services that no longer exist.
• Photos and videos. Film prints, negatives, early digital cameras, every phone we have owned, and a few camcorder tapes.
• Code and projects. Repositories, zip files of repositories, folders that should have been repositories, and the half-finished tools and games that go with them.
• Websites. Old personal sites, blogs, forum posts, and profile pages, some live, most not.
• Everything else. Chat logs, emails, bookmarks, playlists, save files, and the odd floppy disk.

Some of it is on a shelf. Some of it is on drives in a drawer. Some of it is in cloud accounts under passwords I will have to recover. None of it is in one place, and almost none of it is labeled.

Principles

I am setting a few rules before I start, because without them this turns into a decade of tidying that never finishes.

Capture first, judge later. The first pass is not curation. Everything gets collected and inventoried before anything gets thrown away. Deciding what matters is a separate job, and I do not trust my present self to know what a future reader will find interesting.

Originals over summaries. A scan of the letter beats a note that says “got a letter.” A raw export beats a cleaned-up version. If I rewrite something to make it tidier, I keep the original beside it.

Context is the valuable part. A photo is worth little without the date, the place, and who is in it. A repository is worth little without a line about what it was for and why it stopped. Most of the work in this project is writing those lines down while I still can.

Use formats that outlive software. Plain text, PDF/A, open image formats, and folders with readable names. No format that requires a particular vendor to be alive in forty years.

Bridge the physical and the digital. Every box gets a label and an index file. Every scan records which box it came from. The paper does not go away just because it has been scanned, but it does become findable.

Keep the boring parts. The point is the whole, not the best of. Bank statements are part of the record.

The plan

Roughly in order, though the phases will overlap:

1. Inventory. Walk through every box, drive, account, and shelf. Write down what exists, where it is, and in what condition. Do not organize anything yet.
2. Consolidate. Copy every digital source into one place, with checksums, so nothing is lost to a dying disk while the rest of the work happens.
3. Digitize. Scan paper, photos, and negatives. Capture tapes before the players stop working. Archive websites, including the ones only the Wayback Machine remembers.
4. Organize. A single structure for everything, by year and then by kind, with the same naming scheme for paper and files.
5. Annotate. The slow part. Dates, names, places, and a sentence of context for anything that would be a mystery without one.
6. Preserve. Multiple copies, more than one location, more than one medium, and a schedule for checking that the copies still read.
7. Decide what to show. Only once all of the above is done. Some of this will end up on this site. Most of it will not, but it will exist, and it will be findable.

I expect this to take years. That is fine. The archive does not need to be finished to be useful, it only needs to be better organized than it was last month.

What comes next

Posts in this series will cover the tools and the process as I go: how we are scanning, how the folder structure works, what the index files look like, what broke, and what we decided to keep. If you are doing something similar, or thinking about it, I would like to hear how you approached it.

Start with the inventory. That is what I am doing this week.